“It doesn’t make any difference how beautiful your guess is,” Feynman proclaimed, gesticulating in wide, circular, somewhat flamboyant motions. “It doesn’t make any difference how smart you are, who made the guess, or what his name is. If it disagrees with experiment, it’s wrong. That’s all there is to it.”
Feynman was absolutely right.
I agree with this; Feynman had quite a talent for being succinct. However, the converse of this is not true: just because a theory agrees with experiment is not sufficient to confirm that the theory is right — there’s the possibility that competing theories predict the same individual result in some experiment, especially if the prediction is vague. It’s one reason we like mathematical models, which give us specific predictions.